An Overview of CellML 1.1, a Biological Model Description Language

Abstract

CellML is an XML-based exchange format developed by the University of Auckland in collaboration with Physiome Sciences, Inc. CellML 1.1 has a component-based architecture allowing a modeller to build complex systems of models that expand and reuse previously published models. CellML Metadata is a format for encoding contextual information for a model. CellML 1.1 can be used in conjunction with CellML Metadata to provide a complete description of the structure and underlying mathematics of biological models. A repository of over 200 electrophysiological, mechanical, signal transduction, and metabolic pathway models is available at www.cellml.org.

Graphical simulation environments for modelling and simulation of integrative physiology

“…• CellML [31], which provides a representation of the mathematical relationships of biophysical models at the cell level, • TissueML, which describes models at the tissue level, • FieldML [32], describing spatially and temporally varying fields (such as electric fields) related to cell structure, • AnatML, which deals with physiology at the organ level, and • PhysioML [33], which addresses the organ system and organism level.…”

“…CellML has been suggested as a potential solution platform for cell function and to standardise the interface to other computer programs and probably represents the most developed of the IUPS Physiome 'MLs'. A CellML model contains both the essential model equations, written in ascii/markup, and associated 'metadata' which contains [31]:…”

“…An element of hierarchy is also provided for in CellML via the encapsulation and containment grouping relationships. Simulation tools are available for CellML models, via a CellML application programming interface (API) [31], and graphical tools for the development of CellML models are under development. CML and SBML appear to overlap in scope with some of the IUPS Physiome 'MLs', while PhysioML [33] is somewhat unique in its ability to control the interface display and can perform 'computational steering', which refers to the selection of particular variables which can be altered during computation.…”

Interoperability of Neuroscience Modeling Software: Current Status and Future Directions

“…This motivates the development of domain-specific declarative specification languages that cover only the constructional parts of a model (such as which channels are present on which compartments) and assume that functional knowledge (such as how channels affect membrane potential) is built into the simulator. The distinction is clearly exemplified by the approaches used with CellML (Cuellar et al 2003) and SBML (Systems Biology Marup Language) (Hucka et al 2003). The CellML specification aims to allow complete declarative specification of a model.…”

“…Thus, it is not surprising that numerous projects in computational neuroscience and bioinformatics use XML based description formats (e.g. NeuroML (Goddard et al 2002;Crook et al 2005), CellML (Cuellar et al 2003), ChannelML (http://www.neuroconstruct.org/), MorphML (Crook et al 2007), and SBML (Hucka et al 2003). It is somewhat ironic that XML is the most successful representative of declarative languages as it misses out on the initial goal of human readability.…”